Guide / Feasibility

How to run a medical imaging data feasibility study

A medical imaging data feasibility study converts a buyer requirement into explicit inclusion logic, queries aggregate source inventory, validates a representative technical sample, identifies permission and ground-truth dependencies, and records what is verified versus unresolved before a programme is approved.

Published 30 July 2026 · Reviewed 12 August 2026 · MedCorpora
DefineClinical and technical criteria
QueryAggregate inventory
ValidateEvidence and technical sample
DecideProceed · revise · decline
01

What the programme covers

Feasibility protects both buyer and source organization from expensive processing against an ambiguous or impossible specification. It should identify the limiting requirement, not merely return a large preliminary file count.

02

What a useful specification includes

A defensible request defines the clinical task, source evidence and acceptance criteria before patient-level data moves. The exact fields and thresholds depend on the intended model claim.

  • Clinical task and unit of analysis
  • Inclusion, exclusion and control definitions
  • Required imaging series and acquisition constraints
  • Ground truth and clinical linkage
  • Privacy, geography and permitted-use requirements
  • Acceptance metrics, volume and schedule
03

Quality and validation controls

A representative sample can test series classification, linkage, privacy and label assumptions. Sampling results must not be extrapolated without recording the method and uncertainty.

  • Patient versus study versus series counts
  • Required-field completeness
  • Image and report linkage
  • Ground-truth strength and review burden
  • Permission, privacy and delivery dependencies
04

Availability, rights and delivery

The feasibility record should make the go, revise or decline decision auditable. It does not itself authorize patient-level data release.

Public pages describe a sourcing and engineering capability, not guaranteed ready inventory. Each release remains subject to verified programme inventory, programme-specific authorization, privacy review, technical acceptance and buyer licence terms.

05

Questions, answered directly.

Why not begin with a full archive export?

Aggregate queries and samples can expose specification problems before unnecessary patient-level processing.

What can make a cohort infeasible?

Insufficient eligible cases, missing sequences, weak ground truth, incomplete follow-up, permission limits or an incompatible timeline can each be limiting.

Can the buyer trade one requirement for another?

Yes. A feasibility study should expose trade-offs such as volume versus ground-truth strength or protocol precision.

Is inventory under verification ready to order?

No. It remains a separate status until required source, evidence and authority checks are complete.

Institutional engagement

Define the cohort.